Recently, auxiliary devices for learning have been developed which are used in classrooms or lecture rooms in such a way that an auxiliary device captures images of a textbook using a camera and displays the images on a display such as a screen or the like to allow a lecturer or teacher to explain the learning material, or enlarges or reduces the captured images or recognizes characters of the images and outputs the recognized characters in the form of audio speech, thus allowing learners to see the enlarged or reduced images displayed on the display or to listen to the audio speech.
In such an auxiliary device for learning that captures images of a book (a textbook) and displays them or outputs them in the form of audio speech, it is very important to enhance the accuracy with which a digital capture device can capture the images of the book using a camera without distorting the images.
Typical conventional capture devices are classified into the following types, wherein: an open book is placed on the bottom support surface and images of the book are captured using a camera above it; and, as proposed in Korean Patent Registration No. 0765371, an open book is placed upside down on a planar glass panel and images of the book are captured using a camera located below the glass panel.
However, in the case of the type which places the open book on the bottom support surface and the images are captured from above, there is a problem of the pages being undesirably turned. Further, this type forces the user to focus the camera according to the distance between the book and the camera (if the camera has an automatic focusing function, focusing the camera takes some time) and to adjust the position of the book on the bottom support surface.
In the conventional type wherein the open book is placed on the bottom support surface and the images are captured from above or wherein the open book is placed upside down on the planar glass panel and the images are captured from below, images around the medial binding portion of the book may not be captured or may be distorted, because the binding portion of the open book is typically round.
If the images of the book captured by the camera are distorted, not only may the quality of the images of the book that are displayed on a display of an output unit suffer, but there also are many errors for an OCR (Optical Character Reader) engine of the output unit to recognize characters of the images of the book. Moreover, when a TTS (Text-To-Speech) engine converts the characters recognized by the OCR engine into an audio composition signal and outputs the signal, more errors may be produced. Also, it takes a lot of time and effort when the user edits the characters of the distorted images to output the correct audio signal.
In an effort to prevent the captured images of the book from becoming distorted, a digital capture device was proposed in Korean Patent Registration No. 0769107 which was filed by the applicant of the present invention, entitled “Electrical teaching desk with book scanner for scanning books using digital camera”. This digital capture device includes a table which has a reverse ‘V’-shape so that both inner pages of a book come into contact with the table, and a digital camera which captures images of the inner surface of the table below the table.
In this technique introduced by the applicant of the present invention, the reverse ‘V’-shaped table prevents an image captured by the digital camera from becoming distorted. However, as shown in the specification and drawings of this patent document, a lamp which irradiates the inner surface of the table is disposed around the edge of the table. Thus, light cannot be uniformly applied to the inner surface of the table.
Therefore, although there is no distortion of the captured image, the brightness of the image may not be uniform, for example, there may be an excessively bright or dark portion or a portion of the image may be blurred. Such a case may result in the output unit erroneously recognizing the characters of the image and outputting them erroneously in the form of audio speech.
Furthermore, because an LED of the lamp is disposed above the camera, a portion of light emitted from the LED may possibly enter the lens of the camera. If this happens, the quality of the captured image deteriorates, causing an error in recognizing characters of the captured image and outputting them in the form of audio speech.
In addition, when backlight enters the casing from the outside, rather than when emitting the light from the lamp in a casing, the quality of the captured image also deteriorates. Hence, it is needed to prevent backlight from entering the casing. However, the conventional technique registered for the patent has no separate means for blocking backlight from entering the casing through the transparent table. In other words, there are many different kinds of books, and a book placed on the table may not cover the entirety of the table, causing the possibility of backlight entering the casing through the portion of the table that is not covered by the book.
Moreover, to prevent the LED, the camera, a power supply, etc. installed in the casing from malfunctioning and having a shortened life, heat generated by these components must be rapidly dissipated. However, it is difficult for the structure of the conventional technique to dissipate generated heat, because the bottom, front, rear, left and right panels of the casing are closed to prevent backlight from entering the casing.